This collaborative, multi-PI project seeks to gain a detailed understanding of how a potent oncogene, CDK8, and its highly homologous paralog, CDK19, control AML cell growth. We discovered that inhibition of CDK8 and CDK19 kinase activity selectively kills acute myeloid leukemia (AML) cells in culture and in mouse models of AML. As Mediator-associated kinases, CDK8 and CDK19 are poised to broadly regulate gene expression. (Mediator is a multi-subunit complex that regulates genome-wide expression of RNA polymerase II transcripts.) Remarkably, however, global gene expression profiling in multiple cell types indicate that inhibition of CDK8 and CDK19 kinase activity has limited effects on transcription. These results suggest distinct roles for the Mediator kinases CDK8 and CDK19 (i.e. beyond transcription) in the regulation of AML cell growth. A key element of this project is the ability to rapidly and selectively inhibit CDK8 and CDK19 kinase activity in cells. Over the past four years, we have rigorously examined the natural product cortistatin A (CA) for its ability to selectively inhibit Mediator kinases in vitro, in cells, and in vivo. As detailed inthe proposal (and in a forthcoming publication), CA is an unusually potent and selective inhibitor of CDK8 and CDK19 (e.g. selective even at 100-times the IC50 dose). With this first-of-its-kind reagent, we can address some fundamental and long-standing questions about the biological and cancer-relevant roles of Mediator kinases. Importantly, these questions will be addressed in the therapeutically relevant AML cell line MOLM- 14, recently shown to require Mediator kinase function for proliferation and survival. A first critical step toward understanding the biological role of any kinase is to identify its substrates. Few Mediator kinase substrates are known; among these, phosphorylation by CDK8 (no CDK19 substrates are known) triggers protein degradation in some cases. Therefore, in Aim 1A we propose to complete quantitative proteomics experiments (SILAC), including phosphoproteomics and whole proteome analyses, to identify Mediator kinase substrates and determine how Mediator kinase inhibition affects the MOLM-14 proteome. Another major unanswered question related to Mediator kinases is how/whether they impact cell metabolism. Metabolic changes are fundamentally important for oncogenesis and cell differentiation, and CDK8 is known to regulate each of these processes. Moreover, CDK8 has ancient links to metabolic regulation and nutrient sensing in model organisms, suggesting human CDK8 (and perhaps CDK19) may function more broadly as a metabolic regulator. To begin to seriously address this question, we will complete global, untargeted metabolomics experiments (in collaboration with Metabolon, Inc.) in CA-treated vs. untreated MOLM-14 cells (Aim 1B). Follow-up studies will begin to build upon our results in Aim 1A and Aim 1B; however, we anticipate that many key discoveries will result from this exploratory R21 project that will extend well beyond the 2-year time frame.